Tissue-engineering strategies for cartilage repair are attractive for their potential to promote functional restoration of the damaged tissue. Restoration of cartilage "function is a poorly defined and challenging design goal, however, as cartilage has important mechanical, biochemical and biological behaviors that are difficult to simultaneously target with any one strategy. We propose to design a genetically engineered scaffold for restoration of tissue function specific to articular cartilage and the meniscus, which includes the creation of a novel set of definitions of a "successful outcome". Elastin-like polypeptides (ELP) are useful for genetically engineering biocompatible materials with a broad array of physical behaviors through genetic encoding of amino acid sequence, molecular weight, and sites for controlled cross linking. ELPs are also attractive as a versatile scaffold for cartilage repair, as they are native to musculoskeletal tissues and present no antigenic response and may be designed to polymerize in situ to form an integrated 3D scaffold. In this project, we will engineer and optimize cross linked ELP hydrogel scaffolds for functional repair of two different tissues - articular cartilage and the knee meniscus. We will also explore two distinct approaches to cross linking ELPs - an enzymatic cross linking system and a photo-initiated cross linking system. Finally, we will develop a novel framework for the rational design of scaffolds for cartilage repair. We plan to define"outcomes for each scaffold based on quantifiable parameters of important mechanical, diffusion and degradation behaviors, as well as repair tissue biochemistry, histochemical appearance, and cell proliferation. Select parameters will be determined from experiments conducted with the ELP hydrogel scaffolds in vitro (Specific Aims 1 and 2) or in vivo (Specific Aim 3). Neural network models will be used to construct relationships between genetically encoded features of ELP scaffold composition and structure, and variables describing "outcome", to identify patterns amongst these complex and dissimilar variables (Specific Aims 1and 2). This effort will yield novel information on how measured parameters associate to define a particular "outcome", towards the goal of defining scaffold success across physical, biochemical and biological boundaries. The results of this effort will be: (1) one or more cross linked ELPs that are suitable for repair of cartilage and meniscal defects; (2) a framework for the rational quantitation and interpretation of results from tissue repair strategies; and (3) an extensively trained network model useful for the custom-design of an ELP based scaffold for a variety of cartilaginous tissue repair scenarios.